HAL Id: hal-02088966
https://hal.archives-ouvertes.fr/hal-02088966v2
Submitted on 12 Mar 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
BGK and Fokker-Planck models of the Boltzmann
equation for gases with discrete levels of vibrational
energy
J. Mathiaud, Luc Mieussens
To cite this version:
J. Mathiaud, Luc Mieussens. BGK and Fokker-Planck models of the Boltzmann equation for gases with discrete levels of vibrational energy. Journal of Statistical Physics, Springer Verlag, In press, �10.1007/s10955-020-02490-7�. �hal-02088966v2�
BGK and Fokker-Planck models of the Boltzmann equation for gases with discrete levels of vibrational energy
J. Mathiaud1,2, L. Mieussens2
1CEA-CESTA
15 avenue des sablières - CS 60001 33116 Le Barp Cedex, France
(julien.mathiaud@cea.fr)
2Univ. Bordeaux, Bordeaux INP, CNRS, IMB, UMR 5251, F-33400 Talence, France.
(Luc.Mieussens@math.u-bordeaux.fr)
Abstract
We propose two models of the Boltzmann equation (BGK and Fokker-Planck models) for rarefied flows of diatomic gases in vibrational non-equilibrium. These models take into account the discrete repartition of vibration energy modes, which is required for high temperature flows, like for atmospheric re-entry problems. We prove that these models satisfy conservation and entropy properties (H-theorem), and we derive their corresponding compressible Navier-Stokes asymptotics.
Keywords: Fokker-Planck model, BGK model, H-theorem, Rarefied Gas Dynamics, vibrational molecules
1
Introduction
Numerical simulation of atmospheric reentry flows requires to solve the Boltzmann equation of Rarefied Gas Dynamics. The standard method to do so is the Direct Simulation Monte Carlo (DSMC) method [1, 2], which is a particle stochastic method. However, it is sometimes interesting to have alternative numerical methods, like, for instance, methods based on a direct discretization of the Boltzmann equation (see [3]). This is hardly possible for the full Boltzmann equation (except for monatomic gases, see [4]), since this is still much too computationally expensive for real gases. But BGK like model equations [5] are very well suited for such deterministic codes: indeed, their complexity can be reduced by the well known reduced distribution technique [6], which leads to intermediate models between the full Boltzmann equation and moment models [7]. The Fokker-Planck model [8] is another model Boltzmann equation that can give very efficient stochastic particle methods, see [9].
These model equations have already been extended to polyatomic gases, so that they can take into account the internal energy of rotation of gas molecules. They contains correction terms that lead to correct transport coefficients: the ESBGK or Shakhov’s models [10, 11, 12], and the cubic Fokker-Planck and ES-Fokker-Planck [9, 13, 14, 15].
For high temperature flows, like in space reentry problems, the vibrational energy of molecules is activated, and has a significant influence on energy transfers in the gas flow. It is therefore interesting to extend the model equations to take this vibrational modes into account. Several
extended BGK models have been recently proposed to do so, for instance [16, 17, 18, 19], and a recent Fokker-Planck model has been proposed earlier in [13].
All these models assume a continuous vibrational energy repartition. However, while transitional and rotational energies in air can be considered as continuous for temperature larger than 1K and 10K, respectively, vibrational energy can be considered as continuous only for much larger temperatures (2000K for oxygen and 3300K for nitrogen). For flows up to 3000K around reentry vehicles, the discrete levels of vibrational energy must be used [20]. It seems that that the only BGK model that allows for this discrete repartition is the model of Morse [21].
In this paper, we consider a simpler version of this Morse BGK model for vibrating gases that allows for a discrete vibrational energy. We show that the complexity of this model can be reduced with the reduced distribution technique so that the discrete vibrational energy is eliminated. What is new here is that this construction allows us to prove that the corresponding reduced model satisfies the H-theorem. Moreover, the model is shown to give macroscopic Euler and Navier-Stokes equations in the dense regime, with temperature dependent heat capacities, as expected. This means that the reduced model is a good candidate for its implementation in a deterministic simulation code. Note that with this reduction, only higher order moments with respect to the vibration energy variable are lost: the macroscopic quantities of interest like pressure, temperature, and heat flux, are the same as in the non-reduced model. Moreover, since the reduced variable is not the velocity, this reduction does not require any assumption or special geometries.
An equivalent reduced Fokker-Planck model is also proposed, that has the same properties. However, this model is not based on a non-reduced model, since we are not able so far to define diffusion process for the discrete vibrational energy. Up to our knowledge, this is the first time such a Fokker-Planck model for vibration energy is proposed.
Our paper is organized as follows. In section 2, we present the kinetic description of a gas with vibrating molecules, and we discuss the mathematical properties of the reduced distributions that will be used for our models. Our BGK and Fokker-Planck models are presented in sections 3 and 4, respectively. In section 5, the hydrodynamic limits of our models, obtained by a Chapman-Enskog procedure, are discussed. Section 6 gives some perspectives of this work.
2
Kinetic description of a vibrating diatomic gas
2.1 Distribution function and local equilibrium
We consider a diatomic gas. We define f (t, x, v, ε, i) the mass density of molecules with position x, velocity v, internal energy ε, and in the ith vibrational energy level, such that the corresponding vibrational energy is iRT0, where T0 = hν/k is a characteristic vibrational temperature of the
molecule (h and k are the Planck and Boltzmann constant, while ν is the fundamental vibrational frequency of the molecule).
The corresponding local equilibrium distribution is defined by (see [1])
Mvib[f ](v, ε, i) = ρ √ 2πRT3 1 − e−T0/T RT exp − 1 2|u − v|2+ ε + iRT0 RT ! . (1)
Here, ρ is the mass density of the gas, T its temperature of equilibrium and u its mean velocity, defined below.
2.2 Moments and entropy
The macroscopic quantities are defined by moments of f as follows: ρ = hf iv,ε,i, ρu = hvf iv,ε,i, ρe = 1
2|v − u| 2+ ε + iRT 0 f v,ε,i , (2)
where we use the notation hφiv,ε,i=P∞
i=0RR φ(t, x, v, ε, i) dvdε for any function φ.
With standard Gaussian integrals and summation formula, it is easy to find that the moments of the equilibrium Mvib[f ] satisfy:
hMvib[f ]iv,ε,i= ρ, hvMvib[f ]iv,ε,i= ρu.
At equilibrium, we can define the following energies of translation, rotation, and vibration ρetr(T ) = (1 2(v − u) 2)M vib[f ] v,ε,i = 3 2ρRT, (3)
ρerot(T ) = hεMvib[f ]iv,ε,i= ρRT, (4)
ρevib(T ) = h(iRT0)Mvib[f ]iv,ε,i= ρ
RT0
eT0/T− 1 =
δ(T )
2 ρRT, (5)
where the number of degrees of freedom of vibrations is δ(T ) = 2T0/T
eT0/T − 1, (6)
which is a non integer and temperature dependent number, while the number of degrees of freedom is 3 for translation and 2 for rotation.
The temperature T is defined so that Mvib[f ] has the same energy as f : (1 2(v − u) 2+ ε + iRT 0)Mvib[f ] v,ε,i = ρe, which gives the non linear implicit definition of T :
e = 5 + δ(T )
2 RT. (7)
Since the function T → e is monotonic, T is uniquely defined by (7). Moreover, note that δ(T ) is necessarily between 0 and 2, which means that vibrations add at most two degrees of freedom.
Finally, the entropy H(f ) of f is defined by H(f ) = hf log f iv,ε,i.
2.3 Reduced distributions
For computational efficiency, it is interesting to define marginal, or reduced, distributions F and G by F (t, x, v, ε) =X i f (t, x, v, ε, i), and G(t, x, v, ε) =X i iRT0f (t, x, v, ε, i).
The macroscopic variables defined by f can be obtained through F and G only, as it is shown in the following proposition by integrating with respect to v and ε and using the definition (2) of the moments.
Proposition 2.1 (Moments of the reduced distributions). The macroscopic variables ρ, u, and e, of f , defined by (2), satisfy ρ = hF iv,ε, ρu = hvF iv,ε, ρe = (1 2(v − u) 2+ ε)F v,ε + hGiv,ε. (8)
where we use the notation hψiv,ε =RR ψ(t, x, v, ε) dvdε for any function ψ.
This reduction procedure can be extended to the entropy functional as follows. First, to simplify the following relations, we use the notation fi(v, ε) for f (v, ε, i). Then, we define the reduced entropy
by H(F, G) = hH(F, G)iv,ε, where H(F, G) = inf f >0 ( X i
filog fi such that
X i fi = F, X i iRT0fi = G ) . (9)
In other words, for a given couple of reduced distributions (F, G), we define the (non reduced) distribution that minimizes the marginal entropyP
ifilog fi among all the distributions that have
the same marginal distributions F and G. Then the reduced entropy is the integral with respect to v and ε of the corresponding marginal entropy.
Now it is possible to represent this reduced entropy as a function of F and G only, as it is shown in the following proposition.
Proposition 2.2 (Entropy). The reduced entropy H(F, G) defined by (9) is H(F, G) = F log(F ) + F log RT0F RT0F + G + G RT0 log G RT0F + G v,ε . (10)
Proof. The set {f > 0 such that P
ifi = F,
P
iiRT0fi = G} is clearly convex, so that we can use
a Lagrangian multiplier approach by finding if there exists a minimum of the function L defined through : L(f, α, β) =X i filog fi− α X i fi− F ! − β X i iRT0fi− G ! ,
where α and β are real numbers and P
ifilog fi is a convex function of f . The functional L is
convex but no longer strictly convex. A minimum of H(F, G) necessarily satisfies ∂L∂f = 0, and it is easy to deduce that f can be written fi(v, ε) = A exp (−iBRT0), where A := A(v, ε) and
B := B(v, ε) are functions that are still to be determined. The linear constraints give:
F =X i fi= A 1 − exp (−BRT0) , G =X i iRT0fi= ART0exp (−BRT0) (1 − exp (−BRT0))2 ,
where we have used the property iRT0fi = −dfdBi that comes from fi = A exp (−iBRT0). Solving
this linear system gives
A = RT0F 2 RT0F + G , B = 1 RT0 log 1 +RT0F G . so that H(F, G) = F log(F ) + F log RT0F RT0F + G + G RT0 log G RT0F + G . (11)
A final integration with respect to v and ε gives the final result.
The following proposition gives useful differential properties of the reduced entropy functional. Proposition 2.3 (Properties of H). 1. The partial derivatives of H computed at (F, G) are:
D1H(F, G) = 1 + log RT0F2 RT0F + G , D2H(F, G) = 1 RT0 log G RT0F + G . (12) 2. We denote by H = D 11H(F,G) D12H(F,G) D12H(F,G) D22H(F,G)
the Hessian matrix of H. The second order derivatives are: D11H(F, G) = 2 F − RT0 RT0F + G , D12H(F, G) = − 1 RT0F + G , D21H(F, G) = D12H(F, G), D22H(F, G) = F G(RT0F + G) , and we have F D11H(F, G) + GD21H(F, G) = 1, F D12H(F, G) + GD22H(F, G) = 0. (13)
3. The function (F, G) 7→ H(F, G) is convex.
Proof. Points 1 and 2 are given by direct calculations. For point 3, note that the determinant of the Hessian matrix H, which is det H = G(RT01F +G) is positive. Moreover, its trace is positive too, so that the Hessian matrix is positive definite, and hence the function H is convex.
Now, we want to use this reduced entropy to define the corresponding reduced equilibrium. This is done by computing the minimum of the reduced entropy among all the reduced distributions (F1, G1) that have the same moments as (F, G), as it is stated in the following proposition.
Proposition 2.4 (Reduced equilibrium). Let (F, G) be a couple of reduced distributions and ρ, ρu, and ρe its moments as defined by (8). Let S be the convex set defined by
S = (
(F1, G1) such that hF1iv,ε = ρ, hvF1iv,ε= ρu,
(1 2|v| 2+ ε)F 1+ G1 v,ε = ρe ) .
1. The minimum of H on S is obtained for the couple (Mvib[F, G], evib(T )Mvib[F, G]) with Mvib[F, G] = ρ √ 2πRT3 exp −|v − u| 2 2RT 1 RT exp − ε RT (14) where evib(T ) is the equilibrium vibrational energy defined by (5) and ρ, u, T depend on F and
G through the definition of the moments. 2. For every (F1, G1) in S, we have
D1H(F1, G1)(Mvib[F, G] − F1) + D2H(F1, G1)(evib(T )Mvib[F, G] − G1)
≤ H(Mvib[F, G], evib(T )Mvib[F, G]) − H(F1, G1) ≤ 0.
Proof. First, the set S is clearly convex, and it is non empty, since it is easy to see that (Mvib], evib(T )Mvib)
realizes the moments ρ, ρu, and ρe, and hence belongs to S. Now, we define the following Lagrangian L(F1, G1, α, β, γ) = hH(F1, G1)iv,ε− α(hF1iv,ε− ρ) − β · (hvF1iv,ε− ρu) − γ (1 2|v| 2+ ε)F 1+ G1 v,ε − ρe !
for every positive (F1, G1), α ∈ R, β ∈ R3, γ ∈ R. The reduced entropy can reach a minimum of
S when L has its first derivatives equal to zero: it is a minimum if it is unique . Such a point, denoted by (F1, G1, α, β, γ) for the moment, is characterised by the fact that the partial derivatives
of L vanish at (F1, G1, α, β, γ). This gives the following relations (using the cancellation of the L
derivatives in F1, G1, α, β, γ respectively) D1H(F1, G1) = α + β · v + γ 1 2|v| 2, (15) D2H(F1, G1) = γ, (16) hF1iv,ε− ρ = 0, (17) hvF1iv,ε− ρu = 0, (18) (1 2|v| 2+ ε)F 1+ G1 v,ε − ρe = 0, (19)
where D1H and D2H are defined in (12). For instance first relation comes from the fact that the
derivative with respect to F1 satisfies for every δF1
∂F1L(F1, G1, α, β, γ)(δF1) = h(D1H(F1, G1) − (α + β · v + γ( 1 2|v| 2+ ε)))δF 1iv,ε,
It is true for all δF1 leading to the relation 15.
Now Combining equations (15) and (16), one gets that there exists four real numbers A, B, C, D and one vector E ∈ R3, independent of v and ε, such that:
F1 = A exp E · v + B|v|2+ Cε ,
where B and C are necessarily non positive to ensure the integrability of F1 and G1. It is then
standard to use equations (17)–(19) to get F1= Mvib(F, G) and G1 = evib(T )Mvib(F, G).
Finally point 2 is a direct consequence of the convexity of H and of the minimization property.
3
A BGK model with vibrations
With the local equilibrium Mvib[f ] defined in (1), it is easy to derive the following BGK model:
∂tf + v · ∇f =
1
τ(Mvib[f ] − f ). (20)
The macroscopic parameters ρ, u, and T are defined through the moments ρ, ρu and ρe of f (see (2)). Like in the BGK model for monoatomic gases, it will be shown that the relaxation time of this BGK model is τ = µ/p, where p = ρRT is the pressure and µ the viscosity, that can be temperature dependent.
Now we have the following properties.
Property 3.1. • Conservation: for BGK model (20) the mass, momentum and total energy are conserved: ∂t * 1 v 1 2|v| 2 f + v,ε,i + ∇x· * v 1 v 1 2|v| 2 f + v,ε,i = 0.
• H-theorem: for the entropy H(f ) = hf log f iv,ε,i, we have
∂tH(f ) + ∇x· hvf log f iv,ε,i=
1
τ h(Mvib[f ] − f ) log f iv,ε,i≤ 0.
The proof relies on standard arguments (definition of Mvib[f ] and convexity of x log x) and is left to the reader.
3.1 A reduced BGK model with vibrations
For computational reasons, it is interesting to reduce the complexity of model (20) by using the usual reduced distribution technique [22]. We define the reduced distributions
F =X
i
f (t, x, v, ε, i), and G =X
i
iRT0f (t, x, v, ε, i),
and by summation of (20) on i we get the following closed system of two reduced equations: ∂tF + v · ∇xF = 1 τ (Mvib[F, G] − F ) , (21) ∂tG + v · ∇xG = 1 τ( δ(T ) 2 RT Mvib[F, G] − G) , (22)
where the reduced Maxwellian is Mvib[F, G] = ρ √ 2πRT3 exp −|v − u| 2 2RT 1 RT exp − ε RT ,
and the macroscopic quantities are defined by ρ = hF iv,ε, ρu = hvF iv,ε, ρe = (1 2(v − u) 2+ ε)F v,ε + hGiv,ε, (23)
and T is still defined by (7) which implies that T depends both on F and G: to avoid the heavy notation T [F, G], it will still be denoted by T in the following.
Note that this model can easily be reduced once again to eliminate the rotational energy variable. This gives a reduced system of three BGK equations, with three distributions.
It is interesting to compare our new model to the work of [23] and [19]: in these recent papers, the authors also proposed, independently, BGK and ES-BGK models for temperature dependent δ(T ), like in the case of vibrational energy. However, they are not based on an underlying discrete vibrational energy partition, and the authors are not able to prove any H-theorem. Only a local entropy dissipation can be proved. The advantage of our new approach is that the reduced model, which is continuous in energy too, inherits the entropy property from the non-reduced model, and hence a H-theorem, as it is shown below.
3.2 Properties of the reduced model
System (21–22) naturally satisfies local conservation laws of mass, momentum, and energy. More-over, the H-theorem holds with the reduced entropy H(F, G) as defined in (9). Indeed, we have the
Proposition 3.1. The reduced BGK system (21–22) satisfies the H-theorem ∂tH(F, G) + ∇x· hvH(F, G)iv,ε≤ 0,
where H(F, G) is the reduced entropy defined in (9). Proof. By differentiation we get
∂tH(F, G) + ∇x· hvH(F, G)iv,ε = hD1H(F, G)(∂tF + v∇xF ) + D2H(F, G)(∂tG + v∇xG)iv,ε = 1 τ D1H(F, G)(Mvib[F, G] − F ) + D2H(F, G)( δ(T ) 2 RT Mvib[F, G] − G) v,ε ≤ 0
where we have used (21–22) to replace the transport terms by relaxation ones, and point 5 of proposition 2.4 to obtain the inequality.
4
A Fokker-Planck model with vibrations
It is difficult to derive a Fokker-Planck model for the distribution function f with discrete energy levels. We find it easier to directly derive a reduced model, by analogy with the reduced BGK model (21–22) and by using our previous work [15] on a Fokker-Planck model for polyatomic gases. We remind that the original Fokker-Planck model for monoatomic gas can be derived from the Boltzmann collision operator under the assumption of small velocity changes through collisions and additional equilibrium assumptions (see [8]). In practice, the agreement of this model with the Boltzmann equation is observed even when the gas is far from equilibrium (see [9], for instance).
4.1 A reduced Fokker-Planck model with vibrations
First, we remind the Fokker-Planck model for a diatomic gas (without vibrations) obtained in [15]:
∂tf + v · ∇xf = D(f ), (24)
where f = f (t, x, v, ε) and the collision operator is D(f ) = 1
τ ∇v· (v − u)f + RT ∇vf + 2∂ε(εf + RT ε∂εf ) , where the macroscopic values are
ρ = hf iv,ε , ρu = hf viv,ε , ρe = f 1 2(v − u) 2+ ε v,ε = 5 2ρRT.
The internal energy ε can be eliminated by the reduction technique (integration w.r.t dε and εdε) to get
∂tF + v · ∇xF = D1(F , G),
∂tG + v · ∇xG = D2(F , G),
with the collision operators DF(F , G) = 1 τ∇v· (v − u)F + RT ∇vF, DG(F , G) = 1 τ∇v· (v − u)G + RT ∇vG + 2 τ (RT F − G) .
Note that the two velocity drift-diffusion terms in the two previous equations have exactly the same structure as the one in the non-reduced model (24). However, it is interesting to note that the energy drift-diffusion term of (24) gives, after reduction, a relaxation operator in the G equation. Moreover by reducing the model we lose some moments of initial distribution functions (higher moments in internal energy notably) but we are still able to capture energies and fluxes which are generally the main quantities of interest.
By analogy, now we propose the following reduced Fokker-Planck model for a diatomic gas with vibrations. Note that now, the model is still with variables x, v, and ε: only the discrete energy levels i are eliminated. This model is
∂tF + v · ∇xF = DF(F, G), (25) ∂tG + v · ∇xG = DG(F, G), (26) with DF(F, G) = 1 τ ∇v· (v − u)F + RT ∇vF + 2∂ε(εF + RT ε∂εF ) , DG(F, G) = 1 τ ∇v· (v − u)G + RT ∇vG + 2∂ε(εG + RT ε∂εG) + 2 τ (evib(T )F − G) , (27)
where the macroscopic values are defined as in (23) and (7). Again, note that the temperature T depends on F and G.
Note that we do not derive this reduced Fokker-Planck model directly from a model with discrete vibrational energy as for the BGK model, since we are not able so far to define a discrete diffusion operator. As mentioned above, this model is obtained by analogy with the Fokker-Plank model proposed for polyatomic gases. Its derivation from reduction of a discrete in energy Fokker-Plank model will be studied in a future work.
4.2 Properties of the reduced model
Using direct calculations and dissipation properties as in [15] we can prove the following proposition. Proposition 4.1. The collision operator conserves the mass, momentum, and energy:
h(1, v)DF(F, G)iv,ε = 0 and (1 2|v| 2+ ε)D F(F, G) + DG(F, G) v,ε = 0, the reduced entropy H(F, G) satisfies the H-theorem:
∂tH(F, G) + ∇x· hvH(F, G)iv,ε= D(F, G) ≤ 0,
and we have the equilibrium property
(DF(F, G) = 0 and DG(F, G) = 0) ⇔ (F = Mvib[F, G] and G = evib(T )Mvib[F, G]).
Proof. The conservation property is the consequence of direct integration of (27). The equilibrium property can be proved as follows. First, note that the Maxwellian Mvib[F, G] can be written as
Mvib[F, G] = ρ (2π)3/2(RT )5/2exp − 1 2 v − u 2ε T Ω−1v − u 2ε ! , with Ω = RT 0 0 2εRT
. To shorten the notations, Mvib[F, G] will be simply denoted by Mvib
below, and evib(T ) will be simply denoted by evib as well. Then the collision operators can be
written in the compact form DF(F, G) = 1 τ∇v,ε· ΩMvib∇v,ε F Mvib , DG(F, G) = 1 τ∇v,ε· ΩMvib∇v,ε G Mvib + 2 τ (evibF − G) . Then an integration by part gives the following identity for DF(F, G):
DF(F, G) F Mvib v,ε = −1 τ * ∇v,ε F Mvib T ΩMvib∇v,ε F Mvib + v,ε .
Consequently, if DF(F, G) = 0, since the integrand in the previous relation is a definite positive form, the gradient is necessarily zero, and hence F = Mvib. For the equilibrium property of G, the proof is a bit more complicated. First, we have
DG(F, G) G evibMvib v,ε = − 1 τ evib * ∇v,ε G Mvib T ΩMvib∇v,ε G Mvib + v,ε + 2 τ (evibF − G) G evibMvib v,ε . Consequently, if DG(F, G) = 0, and since F = Mvib, we have
1 evib * ∇v,ε G Mvib T ΩMvib∇v,ε G Mvib + v,ε = 2 τ (evibMvib− G) G evibMvib v,ε = −2 τ (evibMvib− G)2 1 evibMvib v,ε + 2 τ hevibMvib− Giv,ε ≤ 2 τ hevibMvib− Giv,ε= 2
which comes from (8) and F = Mvib. Therefore, we obtain 1 evib * ∇v,ε G Mvib T ΩMvib∇v,ε G Mvib + v,ε ≤ 0,
and again this gives G = evibMvib, which concludes the proof of the equilibrium property.
The proof of the H-theorem is much longer. First, by differentiation one gets that the quantity D(F, G) = ∂tH(F, G) + ∇x· hvH(F, G)iv,ε satisfies:
D(F, G) = hD1H(F, G)(∂tF + v · ∇xF ) + D2H(F, G)(∂tG + v · ∇xG)iv,ε
= hD1H(F, G)DF(F, G) + D2H(F, G)DG(F, G)iv,ε, (28)
from (21–22). Then the proof is based on the convexity of H(F, G): while for the BGK model we only used the first derivatives of H, we now use the positive-definiteness of the Hessian matrix of H. To do so we integrate by parts D(F, G) and multiply by τ so that:
τ D(F, G) = − 3 X i=1 h∂vi(F )D11H(F, G) (F (vi− ui) + RT ∂viF )iv,ε −2 h∂ε(F )D11H(F, G) (F ε + RT ε∂εF )iv,ε − 3 X i=1 h∂vi(G)D21H(F, G) (F (vi− ui) + RT ∂viF )iv,ε −2 h∂ε(G)D21H(F, G) (F ε + RT ε∂εF )iv,ε − 3 X i=1 h∂vi(F )D12H(F, G) (G(vi− ui) + RT ∂viG)iv,ε −2 h∂ε(F )D12H(F, G) (Gε + RT ε∂εG)iv,ε − 3 X i=1 h∂vi(G)D22H(F, G) (G(vi− ui) + RT ∂viG)iv,ε −2 h∂ε(G)D22H(F, G) (Gε + RT ε∂εG)iv,ε +2 (evib(T )F − G) 1 RT0 log G RT0F + G v,ε
To use the positive definiteness of the Hessian matrix H of H, we introduce the following vectors: Vi= (F (vi− ui) + RT ∂viF, G(vi− ui) + RT ∂viG)
E = (F ε + RT ε∂εF, Gε + RT ε∂εG),
and we decompose the partial derivatives of F and G in factor of D11F , D22F , D12F as follows:
(∂vi(F ), ∂vi(G)) = 1 RTVi− (F vi− ui RT , G vi− ui RT ) (∂ε(F ), ∂ε(G)) = 1 εE − (F 1 RT, G 1 RT).
This gives τ D(F, G) = 3 X i=1 Fvi− ui RT D11H(F, G) (F (vi− ui) + RT ∂viF ) v,ε +2 F 1 RT D11H(F, G) (F ε + RT ε∂εF ) v,ε + 3 X i=1 Gvi− ui RT D21H(F, G) (F (vi− ui) + RT ∂viF ) v,ε +2 G 1 RT D21H(F, G) (F ε + RT ε∂εF ) v,ε + 3 X i=1 Fvi− ui RT D12H(F, G) (G(vi− ui) + RT ∂viG) v,ε +2 f 1 RT D12H(F, G) (gε + RT ε∂εG) v,ε + 3 X i=1 Gvi− ui RT D22H(F, G) (G(vi− ui) + RT ∂viG) v,ε +2 G 1 RT D22H(F, G) (Gε + RT ε∂εG) v,ε − 3 X i=1 VT i HVi v,ε− 2E T HE v,ε +2 (evib(T )F − G) 1 RT0 log G RT0F + G v,ε
Now this expression can be considerably simplified by using property (13), and we get
τ D(F, G) = 3 X i=1 vi− ui RT (F (vi− ui) + RT ∂viF ) v,ε +2 1 RT (F ε + RT ε∂εF ) v,ε − 3 X i=1 VitHVi− EtHE −2 (evib(T )F − G) 1 RT0 log G RT0F + G v,ε .
to get
τ D(F, G) = 2
RT(ρevib(T ) − hGiv,ε) − 3 X i=1 VitHVi− 2EtHE +2 (evib(T )F − G) 1 RT0 log G RT0F + G v,ε .
The terms with the Hessian are clearly negative, since H is positive definite. Then we have
τ D(F, G) ≤ 2
RT(ρevib(T ) − hGiv,ε) +2 (evib(T )F − G) 1 RT0 log G RT0F + G v,ε . Note that from (8) the first term can be written as
2
RT(ρevib(T ) − hGiv,ε) = 2
RT hevib(T )F − Giv,ε, and can be factorized with the second term to find
τ D(F, G) ≤ 2 (evib(T )F − G) 1 RT0 log G RT0F + G + 1 RT v,ε .
We can now prove that the integrand of the right-hand side is non-positive. Indeed, assume for instance that the first factor is non-positive, that is to say evib(T )F − G ≤ 0. By using evib(T ) =
RT0
eT0/T−1 (see definition (5)), it is now very easy to prove the following relations
evib(T )F − G ≤ 0 ⇔ 1 T0 log G RT0F + G ≥ −R T that is to say the second factor of the integrand is non-negative.
Consequently, we have proved τ D(F, G) ≤ 0, which concludes the proof.
5
Hydrodynamic limits for reduced models
With a convenient nondimensionalization, the relaxation time τ of the reduced BGK model (21)– (22) and the Fokker-Planck model (25)-(26) is replaced by Kn τ , where Kn is the Knudsen number, which can be defined as a ratio between the mean free path and a macroscopic length scale. It is then possible to look for macroscopic models derived from BGK and Fokker-Planck reduced models, in the asymptotic limit of small Knudsen numbers.
For convenience, these models are re-written below in non-dimensional form. The BGK model is ∂tF + v · ∇xF = 1 Kn τ (Mvib[F, G] − F ) , (29) ∂tG + v · ∇xG = 1 Kn τ( δ(T ) 2 T Mvib[F, G] − G) , (30)
where Mvib[F, G] can be defined by (14) with R = 1. Similarly, the relations (3)–(7) between the translational, internal, and total energies and the temperature, have to be read with R = 1 in non-dimensional variables. We remind that T is still a function of F and G. The Fokker-Planck model is ∂tF + v · ∇xF = DF(F, G), (31) ∂tG + v · ∇xG = DG(F, G), (32) with DF(F, G) = 1 Kn τ ∇v· (v − u)F + T ∇vF + 2∂ε(εF + T ε∂εF ) , DG(F, G) = 1 Kn τ ∇v· (v − u)G + T ∇vG + 2∂ε(εG + T ε∂εG) + 2 Kn τ (evib(T )F − G) . (33) 5.1 Euler limit
In this section, we prove the following proposition:
Proposition 5.1. The mass, momentum, and energy densities (ρ, ρu, E = 12ρu2+ ρe) of the solu-tions of the reduced BGK and the Fokker-Planck models satisfy the equasolu-tions
∂tρ + ∇x· ρu = 0,
∂tρu + ∇x· (ρu ⊗ u) + ∇p = O(Kn ),
∂tE + ∇x· (E + p)u = O(Kn ),
(34)
which are the Euler equations, up to O(Kn ). The non-conservative form of these equations is ∂tρ + ∇x· ρu = 0, ρ(∂tu + (u · ∇x)u) + ∇p = O(Kn ), ∂tT + u · ∇xT + T cv(T ) ∇x· u = O(Kn ), (35)
where cv(T ) = dTd e(T ) is the specific heat capacity at constant volume.
Proof. The reduced BGK model (21)–(22) is multiplied by 1, v, and 12|v|2+ ε and integrated with
respect to v and ε, which gives the following conservation laws: ∂tρ + ∇x· ρu = 0,
∂tρu + ∇x· (ρu ⊗ u) + ∇xσ(F ) = 0,
∂tE + ∇x· Eu + ∇x· σ(F )u + ∇x· q(F, G) = 0,
(36)
where σ(F ) = h(v − u) ⊗ (v − u)F iv,εis the stress tensor, and q(F, G) =(v − u)(12|v − u|2+ ε)F
v,ε+
h(v − u)Giv,ε is the heat flux.
When Kn is very small, if all the time and space derivatives of F and G are O(1) with respect to Kn , then (29)–(30) imply F = Mvib[F, G] + O(Kn ) and G = evib(T )Mvib[F, G] + O(Kn ). Then
it is easy to find that σ(F ) = σ(Mvib[F, G]) + O(Kn ) = pI + O(Kn ) , where I is the unit tensor, and q(F, G) = q(Mvib[F, G], evib(T )Mvib[F, G]) + O(Kn ) = O(Kn ), since the heat flux is zero at
equilibrium, which gives the Euler equations (35). The same analysis can be applied for the reduced Fokker-Planck model (31)–(33).
Finally, the non conservative form is readily obtained from the conservative form. Note another formulation of the energy equation that will be useful below:
∂tevib(T ) + u · ∇xevib(T ) +
T e0vib(T ) cv(T )
∇x· u = O(Kn ), (37)
where e0vib(T ) = dTd evib(T ).
5.2 Compressible Navier-Stokes limit
In this section, we shall prove the following proposition:
Proposition 5.2. The moments of the solution of the BGK and Fokker-Planck kinetic models (21)-(22) and (25)-(26) satisfy, up to O(Kn2), the Navier-Stokes equations
∂tρ + ∇ · ρu = 0,
∂tρu + ∇ · (ρu ⊗ u) + ∇p = −∇ · σ,
∂tE + ∇ · (E + p)u = −∇ · q − ∇ · (σu),
(38)
where the shear stress tensor and the heat flux are given by
σ = −µ ∇u + (∇u)T − α∇ · u, and q = −κ∇ · T, (39)
and where the following values of the viscosity and heat transfer coefficients (in dimensional vari-ables) are µ = τ p, and κ = µcp(T ) for BGK, µ = 1 2τ p, and κ = 2 3µcp(T ) for Fokker-Planck, (40)
while the volumic viscosity coefficient is α = cp(T )
cv(T )−1 for both models, and cp(T ) =
d
dT(e(T )+p/ρ) =
cv(T ) + R is the specific heat capacity at constant pressure. Moreover, the corresponding Prandtl
number is
Pr = µcp(T )
κ = 1 for BGK, and
3
2 for Fokker-Planck. (41)
Note that both models do not provide a correct Prandtl number, which can lead to errors for the computation of fluxes in numerical simulations. This is a usual problem with single parameter models like BGK or Fokker-Planck: this can be corrected by a modification of the models like with the ES-BGK or ES-FP approaches, as it has been done for polyatomic gases (see [11, 15] for instance).
5.2.1 Proof for the BGK model
The usual Chapman-Enskog method is applied as follows. We decompose F and G as F = Mvib[F, G] + Kn F1 and G = evib(T )Mvib[F, G] + Kn G1, which gives
Then we have to approximate σ(F1) and q(F1, G1) up to O(Kn ). This is done by using the previous
expansions and (21) and (22) to get
F1= −τ (∂tMvib[F, G] + v · ∇xMvib[F, G]) + O(Kn ),
G1 = −τ (∂tevib(T )Mvib[F, G] + v · ∇xevib(T )Mvib[F, G]) + O(Kn ).
This gives the following approximations
σ(F1) = −τ h(v − u) ⊗ (v − u)(∂tMvib[F, G] + v · ∇xMvib[F, G])iv,ε+ O(Kn ), (42)
and q(F1, G1) = − τ (v − u)(1 2|v − u| 2+ ε)(∂ tMvib[F, G] + v · ∇xMvib[F, G]) v,ε
− τ h(v − u)(∂tevib(T )Mvib[F, G] + v · ∇xevib(T )Mvib[F, G])iv,ε+ O(Kn ).
(43)
Now it is standard to write ∂tMvib[F, G] and ∇xMvib[F, G] as functions of derivatives of ρ, u,
and T , and then to use Euler equations (34) to write time derivatives as functions of the space derivatives only. After some algebra, we get
∂t(Mvib(F, G)) + v · ∇x(Mvib(F, G)) = ρ T52 M0(V )e−J A · ∇T√ T + B : ∇u + O(Kn ), (44) where V = v − u√ T , J = ε T, M0(V ) = 1 (2π)32 exp(−|V | 2 2 ) A = |V | 2 2 + J − 7 2 V, B = V ⊗ V − 1 cv 1 2|V | 2+ J +e 0 vib(T ) cv(T ) I. Then we introduce (44) into (42) to get
σij(F1) = −τ ρTViVjBklM0e−J
V,J∂xluk+ O(Kn ),
where we have used the change of variables (v, ε) 7→ (V, J ) in the integral (the term with A vanishes due to the parity of M0). Then standard Gaussian integrals (see appendix A) give
σ(F1) = −µ ∇u + (∇u)T − α∇ · u I + O(Kn ),
with µ = τ ρT and α = cp
cv − 1, which is the announced result, in a non-dimensional form.
For the heat flux, we use the same technique. First for evib(T )Mvib[F, G] we obtain
∂t(evibMvib(F, G)) + v · ∇x(evibMvib(F, G)) =
ρ T32 M0(V ) ˜ A · ∇T√ T + ˜B : ∇u + O(Kn ), (45) where ˜ A = |V | 2 2 + J − 7 2 + T e0vib(T ) evib V, ˜ B = V ⊗ V − 1 cv 1 2|V | 2+ J +e 0 vib(T ) cv(T ) + T e 0 vib(T ) cv(T )evib I.
Then q(F1, G1) as given in (43) can be reduced to qi(F1, G1) = −τ ρT 1 2|V | 2V iAjM0e−J V,J +ViJ AjM0e−J V,J ! ∂xjT − τ ρDViA˜jM0e−J E V,J∂xjT.
Using again Gaussian integrals , we get
q(F1, G1) = −κ∇xT,
where κ = µcp(T ) with cp(T ) = dTd (e(T ) +ρp) = 72+ e0vib(T ) = 1 + cv(T ) in a non-dimensional form.
5.2.2 Proof for the Fokker-Planck model
Here, we rather use the decomposition F = Mvib(1 + Kn F1) and G = evibMvib(1 + Kn G1), which
gives
σ(F ) = pI + Kn σ(MvibF1) and q(F, G) = Kn q(MvibF1, evibMvibG1),
in which, for clarity, the dependence of Mvib on F and G has been omitted, and the dependence of evib on T as well. Finding F1 and G1 is less simple than for the BGK model: however, the
computations are very close to what is done in the standard monatomic Fokker-Planck model (see [14] for instance), so that we only give the main steps here (see appendix A for details).
First, the decomposition is injected into (33) to get DF(F, G) =
1
τMvibLF(F1) + O(Kn ), DG(F, G) =
1
τevibMvibLG(F1, G1) + O(Kn ), where LF and LG are linear operators defined by
LF(F1) = 1 Mvib ∇v· (T Mvib∇vF1) + ∂ε(2T εMvib∂εF1) , LG(F1, G1) = 1 evibMvib
∇v· (T evibMvib∇vG1) + 2∂ε(T εevibMvib∂εG1) + 2(F1− G1)
.
(46)
Then the Fokker-Planck equations (31)-(32) suggest to look for an approximation of F1 and G1
up to O(Kn ) as solutions of
∂tMvib+ v · ∇xMvib =
1
τMvib(F, G)LF(F1) ∂tevibMvib+ v · ∇xevibMvib=
1
τevibMvib(F, G)LG(F1, G1).
By using (44)-(45), these relations are equivalent, up to another O(Kn ) approximation, to LF(F1) = τ A · ∇T√ T + B : ∇u , and LG(F1, G1) = τ ˜ A · ∇T√ T + ˜B : ∇u , (47)
Now, we rewrite LF(F1) and LG(F1, G1), defined in (46), by using the change of variables V = v−u√ T and G = ε T to get LF(F1) = −V · ∇VF1+ ∇V · (∇VF1) + 2 ((1 − J )∂JF1+ J ∂J JF1) , LG(F1, G1) = LF(G1) + 2(F1− G1).
Then simple calculation of derivatives show that A, B, ˜A, and ˜B satisfy the following properties LF(A) = −3A, LF(B) = −2B,
LG(A, ˜A) = −3 ˜A, LG(B, ˜B) = −2 ˜B.
Therefore, we look for F1 and G1 as solution of (47) under the following form
F1 = aA · ∇T √ T + bB : ∇u and G1 = ˜a ˜A · ∇T √ T + ˜b ˜B : ∇u, and we find ˜a = a = −1/3 and ˜b = b = 1/2.
Finally, using these relations into σ and q and using some Gaussian integrals (see appendix A) give
σ(MvibF1) = −µ ∇u + (∇u)T − α∇ · u I
and q(MvibF1, evibMvibG1) = −κ∇xT,
where α = cp
cv − 1, µ =
τ
2ρT , and κ = 2
3µcp(T ), which is the announced result, in a non-dimensional
form.
6
Conclusion
In this paper, we have proposed to different models (BGK and Fokker-Planck) of the Boltzmann equation that allow for vibrational energy discrete modes. These models satisfy the conservation and entropy property (H-theorem), and the vibration energy variable can be eliminated by the usual reduced distribution function. The low complexity of the reduced BGK model can make it attractive to be implemented in a deterministic code, while the Fokker-Planck model can be easily simulated with a stochastic method.
Of course, since these models are based on a single time relaxation, they cannot allow for multiple relaxation times scales. This is not physically correct, since it is known that the relaxation times for translational, rotational, and vibrational energies are very different. However, standard procedures can be used to extend our model, like the ellipsoidal-statistical approach, already used to correct the Prandtl number of the BGK model [11] and Fokker-Plank models [14, 15].
A
Gaussian integrals and other summation formulas
In this section, we give some integrals and summation formula that are used in the paper. First, we remind the definition of the absolute Maxwellian M0(V ) = 1
(2π)32 exp(−
|V |2
2 ). We
denote by hφi = R
relations (see [24], for instance), written with the Einstein notation: hM0iV = 1, hViVjM0iV = δij, hVi2M0iV = 1, h|V |2M0iV = 3, hViVjVkVlM0iV = δijδkl+ δikδjl+ δilδjk, hVi2Vj2M0iV = 1 + 2 δij hViVj|V |2M0iV = 5 δij, h|V |4M0iV = 15, hViVj|V |4M0iV = 35 δij, h|V |6M0i = 105,
while all the integrals of odd power of V are zero. Note that the first relation of each line implies the other relations of the same line: these relations are given here to improve the readability of the paper. From the previous Gaussian integrals, it can be shown that for any 3 × 3 matrix C, we have
hViVjCklVkVlM0iV = Cij + Cji+ Ciiδij.
Finally, we have also used the following relations: Z +∞ 0 J e−JdJ = Z +∞ 0 e−JdJ = 1, and also +∞ X i=0 e−iT0/T = 1 1 − e−T0/T and +∞ X i=0 ie−iT0/T = e −T0/T (1 − e−T0/T)2.
References
[1] G. A. Bird. Molecular Gas Dynamics and the Direct Simulation of Gas Flows. Oxford Engi-neering Science Series, 2003.
[2] Thomas E. Schwartzentruber Iain D. Boyd. Nonequilibrium Gas Dynamics and Molecular Simulation. Cambridge Aerospace Series. Cambridge University Press, 2017.
[3] G. Dimarco and L. Pareschi. Numerical methods for kinetic equations. Acta Numerica, 23:369–520, 2014.
[4] Luc Mieussens. A survey of deterministic solvers for rarefied flows (invited). AIP Conference Proceedings, 1628(1):943–951, 2014.
[5] E.P. Gross, P.L. Bhatnagar, and M. Krook. A model for collision processes in gases. Physical review, 94(3):511–525, 1954.
[6] C. K. Chu. Kinetic-theoretic description of the formation of a shock wave. Phys. Fluids, 8(1):12, 1965.
[7] H. Struchtrup. Macroscopic Transport Equations for Rarefied Gas Flows Approximation Meth-ods in Kinetic Theory. Interaction of Mechanics and Mathematics. Springer, 2005.
[8] C. Cercignani. The Boltzmann Equation and Its Applications, volume 68. Springer-Verlag, Lectures Series in Mathematics, 1988.
[9] M.H. Gorji, M. Torrilhon, and Patrick Jenny. Fokker-Planck model for computational studies of monatomic rarefied gas flows. Journal of fluid mechanics, 680:574–601, August 2011. [10] Jr. Lowell H. Holway. New statistical models for kinetic theory: Methods of construction.
Physics of Fluids, 9(9):1658–1673, 1966.
[11] P. Andries, P. Le Tallec, J.-P. Perlat, and B. Perthame. The Gaussian-BGK model of boltzmann equation with small prandtl number. Eur. J. Mech. B-Fluids, pages 813–830, 2000.
[12] E. M. Shakhov. Generalization of the Krook relaxation kinetic equation. Izv. Akad. Nauk SSSR. Mekh. Zhidk. Gaza, 1(5):142–145, 1968.
[13] M. Hossein Gorji and Patrick Jenny. A Fokker-Planck based kinetic model for diatomic rarefied gas flows. Physics of fluids, 25(6):062002–, June 2013.
[14] J. Mathiaud and L. Mieussens. A Fokker–Planck model of the Boltzmann equation with correct Prandtl number. Journal of Statistical Physics, 162(2):397–414, Jan 2016.
[15] J. Mathiaud and L. Mieussens. A Fokker–Planck model of the Boltzmann equation with correct Prandtl number for polyatomic gases. Journal of Statistical Physics, 168(5):1031–1055, Sep 2017.
[16] Behnam Rahimi and Henning Struchtrup. Capturing non-equilibrium phenomena in rarefied polyatomic gases: A high-order macroscopic model. Physics of Fluids, 26(5):052001, 2014. [17] Zhao Wang, Hong Yan, Qibing Li, and Kun Xu. Unified gas-kinetic scheme for diatomic
molecular flow with translational, rotational, and vibrational modes. Journal of Computational Physics, 350:237 – 259, 2017.
[18] Takashi Arima, Tommaso Ruggeri, and Masaru Sugiyama. Rational extended thermodynamics of a rarefied polyatomic gas with molecular relaxation processes. Phys. Rev. E, 96:042143, Oct 2017.
[19] S. Kosuge, Hung-Wen Kuo, and Kazuo Aoki. A kinetic model for a polyatomic gas with temperature-dependent specific heats and its application to shock-wave structure. submitted, 2019.
[20] J. D. Anderson. Hypersonic and high-temperature gas dynamics second edition. American Institute of Aeronautics and Astronautics, 2006.
[21] T. F. Morse. Kinetic model for gases with internal degrees of freedom. Phys. Fluids, 7(159), 1964.
[22] A. B. Huang and D. L. Hartley. Nonlinear rarefied couette flow with heat transfer. Phys. Fluids, 11(6):1321, 1968.
[23] C. Baranger, G. Marois, J. Mathé, J. Mathiaud, and L. Mieussens. A BGK model for high temperature rarefied gas flows. Work in progress, 2018.
[24] S. Chapman and T.G. Cowling. The mathematical theory of non-uniform gases. Cambridge University Press, 1970.
